Electronic Measurements Exercises and Assignments

GEERT LANGEREIS

ELECTRONIC
MEASUREMENTS:
EXERCISES AND
ASSIGNMENTS

2
Electronic Measurements: Exercises and Assignments
1st edition
© 2020 Geert Langereis & bookboon.com
ISBN 978-87-403-3268-1
3
ELECTRONIC MEASUREMENTS:
EXERCISES AND ASSIGNMENTS Contents
CONTENTS
1 Measurement Theory 6
2 Measurement Errors 8
3 Electric Currents and Potentials 9
4 Network theory 11
5 Basic Sensor Theory 16
6 Sensor-Actuator Systems 22
7 Signal Conditioning and Sensor Read-out 25
Free eBook on
Learning & Development
By the Chief Learning Officer of McKinsey
Download Now
4
EXERCISES AND ASSIGNMENTS Contents
8 ADC and DAC 31
9 Bus Interfaces 34
10 Assignment 1: The position sensor 35
11 Assignment: Strain gauge transducer 39
12 Answers 44
5
EXERCISES AND ASSIGNMENTS Measurement Theory
1 MEASUREMENT THEORY
Question 1.1 SI Units and conversions

Indicate whether the terms below are a quantity or a unit:
a. Hour
b. Velocity
c. Volt
d. Voltage
e. Intensity of light
f. Liter
g. Newton
h. Radians
The annual average energy consumption of a Dutch household was 3100 kWh in 2015.
The Watt is defined as the number of Joules per second.
i. Convert the annual energy consumption to the derived SI unit for energy
In the field of electrical engineering, we use the quantities potential difference (in Volts),
the electrical current (in Ampères), and the electrical resistance (in Ohms). The Ampère is
an SI unit, unlike the Volt and Ohm. In case we want to express the potential difference
in SI units, we should use 1V = 1 kg m2s-3A-1.
j. Express the Ohm in SI units
Question 1.2 The measurement chain

Thermometers for measuring body temperature are available as electronic devices. Such a
thermometer as shown in Figure 1 starts a measurement after the button is pressed. After
the electronics detects a stable temperature for some seconds, a beep signal is given and the
measured temperature is indicated.
6
EXERCISES AND ASSIGNMENTS Measurement Theory
Figure 1: A fever thermometer
What parts of the measurement chain are available in the thermometer of Figure 1, and
explain:
a. A coupling network
b. A transducer
c. ADC
d. Display
Temperatures are indicated with one decimal. In addition, there is an indicator on the screen
showing whether the temperature is stable or still changing.
e. What type of measurement scale is the temperature indicator?

f. What type of scale is the stability indicator?
7
EXERCISES AND ASSIGNMENTS Measurement Errors
2 MEASUREMENT ERRORS
Question 2.1 The USB adapter

A USB port has a power pin that should give “+5V”. With a multimeter we measure this
potential on five different USB adapters:
4.821 V 4.902 V 4.919 V 4.782 V 4.770 V
a. Calculate the mean

b. Calculate the standard deviation
c. What is the systematic error and what is the random error?
d. Is it correct to represent the potentials with three decimals and four digits?
Question 2.2 Error propagation and representation

Consider two resistors:
• R1 = 10.124kΩ
• R2 = 4.695kΩ
Both have a 5% tolerance
a. What is the correct way of writing the value or R1, representing the 5%
tolerance?
b. What is the correct way of writing the value or R2, representing the 5%
tolerance?
c. What is the correct way of writing the equivalent value or R1 and R2 in series
representing the 5% tolerance? As we will see in chapter 3, the equivalent
resistance of two resistors in series is calculated as R1 + R2.
8
EXERCISES AND ASSIGNMENTS Electric Currents and Potentials
3 ELECTRIC CURRENTS
AND POTENTIALS
Question 3.1 Electric Currents and Potentials

European high voltage networks are based on 380kV. Consider a distance of 50km which is
connected by a cable having a 2.5cm diameter. High voltage cables are made of steel with
a resistivity of 18.0·10-8Ω·m.
a. What is the resistance of the described cable per km?

b. How much power is being dissipated in one such cable over 50km when the
cable carries a current of 500A?
Question 3.2 AC Currents

a. What is the amplitude of a 10V RMS signal?
Question 3.3 Resistors in series and in parallel

Given the circuits of Figure 2, Figure 3, and Figure 4.
Figure 2: A network of resistors
9
EXERCISES AND ASSIGNMENTS Electric Currents and Potentials
a. Calculate the equivalent resistance between point A and B of the circuit of Figure 2
b. Calculate the equivalent resistance between point A and B of the circuit of Figure 3
c. Calculate the equivalent resistance between point A and B of the circuit of
Figure 4
d. In case the resistors of the circuit of Figure 4 are 0.5W, what is the maximum
potential difference that can be placed across point A and B?
Question 3.4 The voltage divider

a. Draw a voltage divider that can make 5V out of a 9V battery. Choose the
theoretical resistors such that the current drawn from the battery is 1mA when
the voltage divider is not loaded.
b. How much deviates the intended 5V when the voltage divider is loaded by a
10kΩ resistor?
10
EXERCISES AND ASSIGNMENTS Network theory
4 NETWORK THEORY
Question 4.1 Kirchhoff’s laws

Given the electronic network of Figure 5.
Figure 5: An electronic network of voltage sources and resistors
a. How many Kirchhoff’s current laws (“junction rules”) or Kirchhoff’s voltage laws
(“loop rules”) are needed to solve this network?
b. What is the magnitude and direction of the current through R3?
www.sylvania.com
We do not reinvent
the wheel we reinvent
light.
Fascinating lighting offers an infinite spectrum of
possibilities: Innovative technologies and new
markets provide both opportunities and challenges.
An environment in which your expertise is in high
demand. Enjoy the supportive working atmosphere
within our global group and benefit from international
career paths. Implement sustainable ideas in close
cooperation with other specialists and contribute to
influencing our future. Come and join us in reinventing
light every day.
Light is OSRAM
11
Question 4.2 Equivalent circuit of a power supply

A power supply is set to 10V. With an ideal multimeter (having no load resistance) we
can verify this. It appears the setting is correct: the multimeter indicates 10.00V. However,
when we connect a light bulb of 100Ω, the supply voltage drops to 9.87V.
a. Draw the Thévenin equivalent circuit of this power supply and indicate the
values of the used elements.
b. Convert the Thévenin equivalent circuit to the Norton equivalent circuit.
c. How much power is dissipated in the light bulb?
Question 4.3 Norton and Thévenin source transformations

Given the electronic networks of Figure 6 and .
Figure 6: A network
Figure 7: A second network
a. Draw the Thévenin equivalence of Figure 6 (between node A and B) and give
the values of the components
b. Convert the Thévenin circuit of the previous question to the Norton equivalent
circuit.
12
c. Draw the Thévenin equivalence of Figure 7 (between node A and B) and give
the values of the components
d. What is the error in % we make when we measure the output voltage between
node A and B of the circuit of Figure 7 with a non-ideal voltmeter that has an
internal resistance of 1kΩ?
Question 4.4 Equivalent circuit of a battery

A battery with the label “9V” is used to power a lightbulb of 100.0Ω. The potential difference
across the lightbulb is measured with an ideal voltmeter and appears to be 8.250V. Because
this appears to be a little bit low for a 9V battery, the potential difference without the
lightbulb is measured as well: this appears to be 8.500V.
a. Draw and calculate the Thévenin equivalent circuit for this battery
Question 4.5 Superposition

Given the network of Figure 8.
Figure 8: A network with two voltage sources
a. Calculate UX of the network in Figure 8 using superposition.

b. Calculate UX by using Kirchhoff’s current law in node UX.
13
Question 4.6 The Wheatstone bridge

Consider the bridge circuit of Figure 9 with the temperature dependent resistor RT. The
voltmeter “V” in the middle can be assumed to be ideal (RVoltmeter = ∞Ω):
360°
.
Figure 9: A resistive Wheatstone bridge
thinking
360°
thinking . 360°
thinking .
Discover the truth at www.deloitte.ca/careers Dis
© Deloitte & Touche LLP and affiliated entities.
Discover the truth at www.deloitte.ca/careers © Deloitte & Touche LLP and affiliated entities.
Deloitte & Touche LLP and affiliated entities.
Discover the truth at www.deloitte.ca/careers

14
Take Ua = 10V, R1 = R2 = R3 = 1kΩ.
a. At room temperature (18°C) the value of RT is equal to 1023Ω. What indicates

the voltmeter?
b. The resistance of the temperature sensor can be modelled as RT(T) = 4.93·107
e-T/27.0, with T in Kelvin (Celsius + 273.15) and RT in Ohm. At what
temperature is the voltmeter at exactly 0V?
c. The value calculated at question a) is known as the differential voltage for the
voltmeter. What is the common voltage of the voltmeter?
d. What is the sensitivity of the bridge circuit in Volt per degree?
Question 4.7 Time dependent circuits

Figure 10 is an RC circuit with a constant voltage source that is switched on at t = 0 sec.
Initially, the capacitor is not charged, so at t = 0 sec the capacitor voltage is UC = 0V.
Figure 10: An RC circuit
a. What is the final potential (after waiting for a long time) across the capacitor
after the switch is closed?
b. What is the time constant?
c. Starting with UC = 0V at t = 0 sec, what is the size of the electric current
directly after closing the switch?
d. What is the final current (after waiting for a long time) after closing the switch
e. What happens with the current and potential if we open the switch once the
capacitor is fully charged?
15
EXERCISES AND ASSIGNMENTS Basic Sensor Theory
5 BASIC SENSOR THEORY
Question 5.1 Terminology

What is the name of the following phenomena:
a. The output of a sensor (electrical quantity) is “a” times the input of the sensor
(another physical quantity). How do we call the factor “a”?
b. A certain sensor is only linear up to a certain input value: for higher inputs, the
output is always the supply voltage. What is the name of this phenomenon?
c. A certain sensor has always a reading that is 5 mV higher than the real value.
What is the name of this difference?
d. After switching on the power supply of a sensor unit, the sensor reading slowly
increases until it reaches the real value. What is the name of this time behavior?
e. A certain distance sensor can measure from 10 cm up to 10 m, but the smallest
difference it can “see” is 5 mm (smaller variations disappear in the sensor noise).
How do we call this boundary of the sensor?
f. A certain temperature sensor appears to be sensitive to temperature as well. How
do we call such an artefact?
g. A pH sensor can measure the acidity of a liquid. Before each measurement,
we have to place it in a known solution with pH = 7 and press a certain reset
button. What is the name of this procedure?
h. To remove background noise, we use two microphones and we measure the
difference between them. How do we call this configuration?
i. The differential amplifier in a Wheatstone bridge needs to amplify differential
signals much more than common signals. What is the name of this amplifier
specification item?
j. A certain sensor can track signals up to 100Hz. How do we call the frequency
range of a sensor?
Question 5.2 The transfer function of a Pt100

Consider a Pt100 temperature sensor with the calibration curve of Figure 11.
16
Figure 11: Calibration curve of a Pt100 temperature sensor
Some raw data is available as well in Table 1.
We will turn your CV into

an opportunity of a lifetime
Do you like cars? Would you like to be a part of a successful brand? Send us your CV on
We will appreciate and reward both your enthusiasm and talent. www.employerforlife.com
Send us your CV. You will be surprised where it can take you.
17
T [°C] R [Ohm]
-13.15 94.851
1.85 100.723
26.85 110.452
61.85 123.952
101.85 139.207
121.85 146.765
271.85 201.979
346.85 228.612
436.85 259.713
501.85 281.594
Table 1: Measurement points form a Pt100 sensor
a. Fit a linear model (a·x+b) for the range 0°C to 100°C

b. What is the sensitivity? (include units)
c. What is the offset? And what does it mean?
d. What is the non-linearity in that temperature range? (Defined as the maximum
deviation divided by the full-scale range, expressed as a percentage)
e. Repeat question a to d for the temperature range 0°C to 500°C
f. Do you expect the Pt100 will drift? Because of what process?
g. What is the reference?
Now consider this Pt100 to be placed in a voltage divider with an ideal 100Ω series resistor
and an ideal 5V supply.
a. What is the output voltage over the range 0°C to 100°C? So what is the
sensitivity ΔV/ΔT?
b. What is the non-linearity over the range 0°C to 100°C of the sensor plus this
simple read-out method?
c. So why is a simple voltage divider not the preferred read-out
mechanism for a Pt100?
18
Question 5.3 A resistive sensor

A strain gauge is a resistive element of which the resistance changes as a function of deflection
or stretch. A home-made strain gauge is tested and the measured resistance is in Figure 12.
Figure 12: Measured calibration curve of a strain gauge
19
Using Excel a line is drawn through the measured points (with the least-square method) to
determine the relation between resistance and the deflection. The used model is R = a·d + b,
assuming a linear relation between the resistance R and deflection d.
a. What is the sensitivity of the sensor? Express this as a number with the
appropriate units.
b. What is the non-linearity? Express this as a number.
c. Given the measured points, is there a suggestion to improve the accuracy of the
sensor?
d. Is this a self-generating or modulating sensor?
Question 5.4 The candle-clock

Before there were mechanical clocks, several systems were used to measure time. One of them
was the use of a candle. Assuming the candle burns at a constant speed in a reproducible
way, we can place a linear scale along the candle. An example is shown in Figure 13.
Figure 13: A measurement setup for a candle-clock
20
With an experiment we want to determine the relation between the length of the burning
candle (l) and the time (t). The used model is l = at + b, assuming a linear relation between
length and time. The constant b has the meaning of the length of the candle at t = 0 sec.
The measured curve is in Figure 14.
a. Draw the best fit for the model (l = a·t + b) in the graph
b. Determine the model parameters a and b from your fitted line. Represent them
in SI units.
c. The measurement points are not exactly on the fitted line: there will be a
random error. Explain how we can express the random error in a number. (So: it
is not about the number itself, but about the procedure)
d. What is the burn speed of the candle?
Figure 14: Measured calibration curve of the candle clock
21
EXERCISES AND ASSIGNMENTS Sensor-Actuator Systems
6 SENSOR-ACTUATOR SYSTEMS
Question 6.1 From sensor to knowledge

a. Explain how differential measurements can help to resolve the sensor issues
called “cross-interference” and give an example.
b. What are “Stimulus-response measurements” and how do they help us in sensor
systems?
Question 6.2 Differential measurements

Consider a sensor with a linear transfer curve y(x) = ax +b to measure x in the quantity y.
The quantity y can be a voltage for example. However, the output voltage y may also have an
additional voltage due to noise, for example an induced voltage due to long connection wires.
AXA Global
Graduate Program
Find out more and apply
22
a. Assume that noise is not correlated to the signal x, and only introduces
fluctuations in the offset b. What would happen if we use two sensors: sensor 1.
is exposed to the signal, and sensor 2 is not exposed. Both sensors pick up the
same amount of noise (because it came from the cables). As an output, we do
not take y(x) from a single sensor, but y2(x) - y1(x0). What happens to the noise?
b. What if the noise is not in the offset b, but in the sensitivity a?
c. How could an electronic circuit look like for measuring the difference between
two parameters y1 and y2?
Question 6.3 A resistive measurement bridge for a strain gauge

We have created a strain gauge by placing metal films on two sides of an insulating beam
as shown in Figure 15. The measured quantity is the deflection x.
Figure 15: Configuration with four strain gauges on a lever
a. Draw a Wheatstone resistive bridge and place all strain-gauge resistors R1, R2, R3
and R4 that you think are needed in it. Explain the configuration
b. Sketch the graph of the response of your bridge. Explain the shape. Is it linear
around x = 0 m?
23
c. All resistors are temperature dependent. How can we minimize the error (cross
sensitivity) in the output signal due to temperature changes as much as possible?
d. The output of a resistive bridge is the difference between the voltage in the left
branch and the right branch: this appears to be from -5mV to +5mV maximally
when the bridge has a power supply of Us = 10V. We want to amplify this
difference to feed it into an AD-Converter with an input range -5V to 5V. What
are the design considerations for the amplifier?
e. For temperature sensors, we sometimes use a 4-wire technique. Is that useful in
this Strain gauge set-up? If, yes: how should it be implemented and how does it
work? If not: why not?
24
EXERCISES AND ASSIGNMENTS Signal Conditioning and Sensor Read-out
7 SIGNAL CONDITIONING
AND SENSOR READ-OUT

The Wheatstone bridge of Figure 16 has an LDR sensor, which stands for “Light Dependent
Resistor”. As the name explains, such a sensor is a resistor of which the value is dependent
on the exposed light.
�e Graduate Programme
I joined MITAS because for Engineers and Geoscientists
I wanted real responsibili� www.discovermitas.com
Maersk.com/Mitas �e G
I joined MITAS because for Engine
I wanted real responsibili� Ma
Month 16
I was a construction Mo
supervisor ina const
I was
the North Sea super
advising and the No
Real work he
helping foremen advis
International
al opportunities
Internationa
�ree wo
work
or placements ssolve problems
Real work he
helping fo
International
Internationaal opportunities
�ree wo
work
or placements ssolve pr
25
Figure 16: A Wheatstone bridge with an LDR resistive light sensor
a. At what resistor value of the LDR will the voltmeter “V” indicate 0V?
b. What is the sensitivity of this bridge? In other words: if the LDR changes 1
Ohm how many Volt will be changed on the voltmeter?
Question 7.2 An LDR in a voltage divider

A typical LDR can have a transfer curve as shown in Figure 17. The light intensities of the
moon, a typical living room and in direct sunlight are indicated in the graph. The light
intensity is in Lux (lumens per square meter). We will use this sensor in an office (indoors)
to detect whether a sunscreen has to go down. We decide that when the light is above
1000 lux, the screen has to go down, and we want to put it up again if the light becomes
below 100 Lux.
26
Figure 17: Transfer curve of a typical LDR
a. The sensor is used for controlling a sun-screen in a home automation

application. What range for the light intensity would you choose?
b. In what range for the resistance would that result?
c. We want to put the sunscreen down if the light intensity is higher than 1000
Lux, and we want to put it up again if the light becomes below 100 Lux. Why
would we take two different thresholds? What would happen of we put it down
at 100 Lux and up at 1000 Lux?
d. Put the sensor in a voltage divider. Propose an appropriate series resistor.
e. How to make the two decision levels?
Question 7.3 Sensor characteristics of an NTC

An NTC temperature sensor can be connected to any circuit (like the AD input of a
microcontroller for example) with the circuit of Figure 18.
27
Figure 18: Method to connect an NTC temperature sensor
a. What determines the (non)linearity between the relation of the temperature with
the output voltage Uout?
b. What is the purpose of the amplifier (the triangle with the A)?
93%
OF MIM STUDENTS ARE
WORKING IN THEIR SECTOR 3 MONTHS
FOLLOWING GRADUATION
MASTER IN MANAGEMENT
• STUDY IN THE CENTER OF MADRID AND TAKE ADVANTAGE OF THE UNIQUE OPPORTUNITIES
Length: 1O MONTHS
THAT THE CAPITAL OF SPAIN OFFERS
Av. Experience: 1 YEAR
• PROPEL YOUR EDUCATION BY EARNING A DOUBLE DEGREE THAT BEST SUITS YOUR
Language: ENGLISH / SPANISH
PROFESSIONAL GOALS
Format: FULL-TIME
• STUDY A SEMESTER ABROAD AND BECOME A GLOBAL CITIZEN WITH THE BEYOND BORDERS
Intakes: SEPT / FEB
EXPERIENCE
5 Specializations #10 WORLDWIDE 55 Nationalities

MASTER IN MANAGEMENT
Personalize your program FINANCIAL TIMES
in class
www.ie.edu/master-management [email protected] Follow us on IE MIM Experience
28
c. At room temperature (18°C), the used NTC is 970Ω. It has a sensitivity of

3%/°C. How much changes Uout when we heat from 18°C up to 20°C? Take
R2 = 1kΩ and Ucc = 5V.
d. The answer from the previous question is probably a small voltage difference,
which means that an Analog to Digital converter needs a very high resolution
to see variation in the single degree range. How can we improve the circuit of
Figure 18 to increase the resolution? (Hint: simply amplifying the signal 10
times it probably not going to work).
e. Resistor R2 appears to be sensitive to temperature as well. Explain what happens
and whether we can remove this effect by calibration? (Explain how or why not)
Question 7.4 Schmitt trigger

We have a signal as shown in Figure 19 of which we want to detect whether it is “high” or
“low”. When using a single decision level, there would be many unintended zero crossings.
Therefore, we decide to make a circuit that has a low-to-high decision level of 1.7V and a
high-to-low transition level of 0.9V.
Figure 19: A noisy signal that needs a Schmitt trigger for high-low detection
Such a circuit is called a Schmitt Trigger and a possible implementation is shown in Figure 20.
29
Figure 20: Implementation of a Schmitt trigger with a single OpAmp
a. How does the Schmitt trigger of Figure 20 work to create the action of Figure 19?
b. What are good values for R1, R2 and R3 to create the action of Figure 19?
30
EXERCISES AND ASSIGNMENTS ADC and DAC
8 ADC AND DAC
Question 8.1 AD conversion

A data acquisition system is used to sample a signal between 0 and 5 Volt. We want to do
this with a resolution of 5mV.
a. How many bits are needed for the AD converter to achieve the required
resolution? Explain how you found this answer?
This signal of interest appears to have no frequencies above 1kHz.
b. What is an appropriate sample frequency? Why?
31
Question 8.2 Measurement of light intensity

The Advanced Photonix PDV-P9004 is an LDR (Light Dependent Resistor) with a transfer
curve as shown in Figure 21.
Figure 21: Transfer curve of the Advanced Photonics PDV-P9004 LDR
We want to measure the change in electrical resistance as a change in voltage in order to

sample with a microcontroller. Assume practical light intensities are between 5 lux and 50
lux, and we are using a 12-bit AD-converter configured for a range of 0V up to 5V.
a. What is the output range of a bare sensor for the given measurement range?
b. Can we express the sensitivity as a single number? If yes: what is the sensitivity?
If not: why not?
c. Give an electric circuit with which the resistor value (and so the measured light
intensity) is converted into a measurable voltage.
d. Give the component values of your circuit (meaning the other resistor values and
the power supply if applicable)
The sensor is used to measure when it is day and night.
e. The smallest change of light we want to measure results into a change in voltage
of 1 mV. Can we detect this with the given AD converter?
f. The day-night period is 24 hours. According to the Nyquist rule we should
measure at least two times per 24 hours. Why should we measure at a much
higher rate in our case?
32
Question 8.3 A temperature measurement

In Figure 22 there are two transfer curves of two different temperature sensors. The left one
is of a Pt100 sensor and the right one of an NTC.
Figure 22: Figure 22Pt100 temperature sensor (left-hand) and an NTC (right-hand)
a. Which of the two sensors is the best choice to measure room temperature? Why?
b. We want to convert the change in resistance into a change in voltage. The
change in voltage has to be sampled with an AD converter between 0V and 5V
with 12 bits. Draw a circuit for the conversion of the resistance change to the
intended voltages and explain the chosen component values.
c. In Figure 22 we can see that the curve of the NTC is not linear. How can we
deal with this?
33
EXERCISES AND ASSIGNMENTS Bus Interfaces
9 BUS INTERFACES
Question 9.1 Bus interfaces

a. Many complex systems like cars and home appliances have moved from a
flat architecture to an architecture based on digital buses. Give at least five
advantages to implement the electronic architecture for complex sensor systems
as electronic digital buses.
b. Industrial buses for production machines are more and more based on Ethernet
protocols. What are the advantages?
In the past 5 years we have drilled around
95,000 km
—that’s more than twice around the world.
Who are we?

We are the world’s leading provider of reservoir characterization,
drilling, production, and processing technologies to the oil and
gas industry.
Who are we looking for?

We offer countless opportunities in the following domains:
n Operations
n Research, Engineering, and Manufacturing
n Geoscience and Petrotechnical
n Commercial and Business
We’re looking for high-energy, self-motivated graduates

with vision and integrity to join our team. What will you be?
careers.slb.com
34
EXERCISES AND ASSIGNMENTS Assignment 1: The position sensor
10 ASSIGNMENT 1: THE
POSITION SENSOR
Equipment needed:
• 1 white A4 paper
• 1 carbon grey pencil
• 1 ruler or protractor
• 2 Paperclips
• A multimeter to measure resistances.
With simple office materials it is relatively easy to make some sensors. For example, a linear
position sample can be made as shown in Figure 23.
Figure 23: Make a position sensor using a carbon pencil
35
• Draw a rectangle of 1 cm x 12 cm with a pencil (these carbon ones) on the paper

• Fill the rectangle with a thick layer of pencil coloring, color in two directions
• Draw a ruler with a grid of 1 cm
• Now fold the paper just above the drawn rectangle
• Put a paperclip at 1 cm and a second paperclip at 11 cm (from metal to metal,
the distance is now 10 cm)
Question 10.1 First measurement – does it work?

Measure the resistance between the two paperclips with a multimeter. This must be a
reasonable number. If it is too high (multimeter out of range), make sure you are using a
conductive carbon pencil, apply a thicker layer and/or check if the paperclips make a good
galvanic contact with the strip drawn.
a. What is your typical resistance for d = 10 cm?
Question 10.2 Calibration curve

a. Now measure the resistances for d = 10 cm, 9 cm, 8 cm, ... 1 cm and make a table
d [cm] R [Ω]
Table 2: Measured resistances at various distances
b. Draw the curve for the resistance R as a function of the deflection d
36
Question 10.3 Analyzing

We should see a dependency of the resistance on the applied distance: we have created a
position sensor. Let’s analyze the behavior as a position sensor:
a. What dependency do you see? Linear, exponential, quadratic?

b. Based on the formula for a resistor, did you expect this?
c. In case you do not like the typical value (for example it is in the Tera-Ohms or
milli-Ohms range), how could you re-design the sensor?
d. Are there values on the curve you do not trust? Why? What could happen there?
e. Assuming the dependency R on d is linear (meaning y = a∙x + b with y the
measured resistance R and x the distance d), can you fit a line and estimate the
slope “a” and the offset “b”? (Excel has several option to do this, but a rough
manual estimation is sufficient too)
f. What does the slope “a” mean in this sensor? Where does it come from?
g. What does the offset “b” mean in this sensor? Where does it come from?
Excellent Economics and Business programmes at:
“The perfect start

of a successful,
international career.”
CLICK HERE
to discover why both socially
and academically the University
of Groningen is one of the best
places for a student to be
www.rug.nl/feb/education
37
Question 10.4 Error estimation

a. Measure R at d = 2 cm (again). Remove the paperclips, shift the paperclips, put
them back on and repeat the measurement at d = 2 cm. Repeat this 5 times to
check the reproducibility of the experiment
b. Do this also for d = 6 cm and d = 10 cm
c. What do you see? Can you see the nature of the error (is it just 10%
fluctuation, or does it change continuously?). What in your setup determines the
error in this setup?
Question 10.5 Repeat calibration experiment

a. Repeat the experiment with the calibration curve of Question 10.2, meaning:
measure R as a function of d again
b. Make a graph
Question 10.6 Analysis

a. Is the slope “a” and the offset “b” still the same?
b. Explain changes. What physical phenomenon in the sensor takes place that
changes the slope and/or the offset?
c. What is the most critical shortcoming in this simple sensor? In other words,
what do we have to improve to make it useful?
38
EXERCISES AND ASSIGNMENTS Assignment: Strain gauge transducer
11 ASSIGNMENT: STRAIN
GAUGE TRANSDUCER
Equipment needed:
• 1 strip of plastic material, typically 1-2 mm thick

• 1 carbon grey pencil
• 1 ruler or protractor
• 2 paperclips
• A multimeter to measure resistances.
A strain gauge sensor is a sensor to measure deformation of a hard material. It can be

made using a carbon film similar to assignment 11. In this assignment we will use it to
measure the curvature N of a beam. Figure 24 shows a solid beam that is bend with a
radius of curvature R.
Figure 24: A length change due to bending a rod
An axial strain H, meaning a relative elongation of the length L, is caused by the bending
and equal to
fL
'y y !(.
L
In short, due to the curving of the beam with a radius R, the unstrained length L becomes
(1+H)L after stretching, where H = y/R with y half of the thickness of the beam. If we place
a resistive strip on one side of the beam, the change in length of this resistive strip will
39
result linearly in a change of resistance because the length L will change. From theory we
know that the length L determines the resistance R as

y*

with ࣁ the resistivity of the material [Ω/m] and A the cross-sectional surface area.
Question 11.1 First measurement - does it work?

a. Make such a sensor by drawing a square of conductive carbon on one side of the
plastic strip. The paperclips can be used to contact the strip in order to measure
the resistance over the full length.
b. What is the sensitivity? This can be in “Ohm per mm bending” for example
c. Is it linear?
d. What is the first thing to improve in this transducer? How?
American online
LIGS University
is currently enrolling in the
Interactive Online BBA, MBA, MSc,
DBA and PhD programs:
▶▶ enroll by September 30th, 2014 and

▶▶ save up to 16% on the tuition!
▶▶ pay in 10 installments / 2 years
▶▶ Interactive Online education
▶▶ visit www.ligsuniversity.com to
find out more!
Note: LIGS University is not accredited by any

nationally recognized accrediting agency listed
by the US Secretary of Education.
More info here.
40
Question 11.2 Read-out electronics

a. How can you turn the change of R (until now probably measured with a
multimeter) into a voltage for sampling with a 0-5V ADC? Consider the range
and the appropriate output levels.
Question 11.3 The strain gauge

The equation

y*

can be used to assess the setup quantitatively. The resistivity U of carbon (graphite) is 3
to 60 × 10-5 Ω∙m. The temperature coefficient is -0.0005 per °C. In a first explorative
experiment with a carbon pencil resistor, resistances were found of 10MΩ to 35MΩ for a 1
cm wide and 10 cm long resistor. This means, that the thickness of a thick pencil drawing
is somewhere between 0.1 and 0.03 nm. From these explorative experiments, we can write:

y* y* y ."0&
5
which gives a tool to design pencil resistors based on their L/W ratio, with L the length
and W the width.
a. Determine your ratio between R and L/W from the experiments

b. Assume we can optimize our carbon layer to U/d = 0.5MΩ. What is the length
of the strain gauge for a width of 2.5cm and a target resistance of 1MΩ?
What should be the shape of the strain gauge: a straight bar or do we have to
meander?
c. Assume the full scale change in resistance due to strain is 1%, how much
temperature variation do we allow for 5% error?
Question 11.4 Bridge set-up

Consider the bridge setup of Figure 25.
41
Figure 25: A resistive bridge set-up (Wheatstone bridge)
a. Calculate or simulate the change in UBridge for a change in ΔR. If you need
numbers, then you can take our target R0 of 0.5MΩ.
b. What is the ratio between the common mode signal on the UBridge pins and the
differential mode signal?
c. What happens to the UBridge relation with ΔR if our strain gauge is not R0 as
intended (and as the other three resistors in the bridge), but if it is 20% higher?
What is the consequence for the amplifier we need to measure UBridge?
Question 11.5 Half bridge setup

Apparently, we have a poor sensor with a lot of process variations, drift and temperature
sensitivity. One way to overcome some common effects is to use a half-bridge. If we make
two “identical” strain gauges, and we place one on a mechanical place with tension as shown
in Figure 26, and one in compression, we have some advantages.
Figure 26: A half-bridge set-up
42
a. Simulate or calculate a bridge with two opposing sensors as shown in the half
bridge. Compare the effect on UBridge (in voltage change) for an applied ΔR. All
other settings (R0, UV) are the same. What is the difference?
b. Now consider a common effect, for example a temperature change. This means
that all R0’s change and ΔR is the same. What happens to the output UBridge?
43
EXERCISES AND ASSIGNMENTS Answers
12 ANSWERS
Question 1.1 SI Units and conversions

a. Unit
b. Quantity
c. Unit
d. Quantity (incorrect expression for potential difference)
e. Quantity
f. Unit
g. Unit
h. Unit
i. 3100 kWh = 3100 kW · 3600 sec = 3.1·106 · 3.6·103 W·s = 11.16·109 J = 11.16 GJ
j. Ohm = Volt/Ampere = 1 kg m2s-3A-1 /A = 1 kg m2s-3A-2
Question 1.2 The measurement chain

a. Yes: the metal tip making an optimum thermal coupling to the human body
b. Yes: there must be a sensor, for example an NTC
c. Yes: since the display is digital, there must be an analog to digital conversion
d. The LCD is a display indicating the temperature value, but the indicator
indicating a stable setting is also part of the display
e. Temperature in degrees of Celsius or Fahrenheit is an interval scale
f. A binary scale
Question 2.1 The USB adapter

a. Add and divide by n = 5:
/"3-, v /"4+- v /"4,4 v /"23- v /"22+

y /"3./
0
44
b. Add squared and divide by n-1:
/"3-, w /"3./? v /"4+- w /"3./? v /"4,4 w /"3./? " "

v/"23- w /"3./? v /"22+ w /"3./?
y 13"04
0w,
c. 4.834V - 5.000V = -166 mV

d. No, the random error is 68.59 mV = 0.07 V, so two decimals is enough
Question 2.2 Error propagation and representation

a. 10.1 Ω 0.5 Ω
b. 4.7 Ω 0.2 Ω
c. Absolute errors add up: 14.8 Ω 0.7 Ω
Question 3.1 Electric Currents and Potentials

a. Resistance of a rod:
,"+ 5 ,+@
y* y* ?y* y ,3"+ 5 ,+HE y +".2&
) ? -"0 5 ,+H?
?
) - )
-
b. Express the dissipated power in terms of the current through the cable:
c.
y 5 y ? y 0++? +".2& 5 0+ y /"1-0
Question 3.2 AC Currents

a. y Rkonp y ,/",
l
8?
45
Question 3.3 Resistors in series and in parallel

? @
?@ y
? v @
?@
a. >?@ y > v
? v @
>?@ A
>?@A y z /./&
>?@ v A
>@A y > v @ v A
b.
>@A ?
>?@A y y 20&
>@A v ?
c. ?@
>?@ y > v y .++&
? v @
R R dR
d. Calculate power dissipated in R1, R2 and R3:
> ,

Ok y Ok 5 Ok y JK 5 JK { +"0
>?@ >?@
46
?@ @ # ?@

Ol y Ol 5 Ol y JK 5 JK { +"0
>?@ >?@
?@ ? # ?@

Om y Om 5 Om y JK 5 JK { +"0
>?@ >?@
Look at similar terms and find that at constant UAB, the powers are: PR1 < PR2 <PR3. So R3
will dissipate most of the power:
p
? ,&

Om y JK ? y JK ?
{ +"0
>?@ ? .++&?
.++&?
JK { +"0 z 1"2,
,&
Question 3.4 The voltage divider

a.
Figure 27: Voltage divider to make 5V out of 9V
Two boundary conditions:
adc
y Ol y y ,
?
?
adc y ]`
> v ?
So:
0
? y y 0&
,
]`
> y ? w , y /&
adc
47
Question 4.1 Kirchhoff’s laws

a. There are 6 nodes (points between two or more components) for the current law:
• One is 0V per definition,
• The two voltage sources eliminate one node each,
• The nodes between R1/Ra and R2/Rb do not have to be evaluated because
series resistors can be taken together as R1+Ra and R2+Rb respectively (they
have the same current).
This leaves only one node for the Kirchhoff’s current law, which is the node
connecting R1, R2 and R3 when the node between Ua, R3 and Rb is taken as the
grounded node.
There are two loops (mazes) for Kirchhoff’s Voltage Law.
b. Apply Kirchhoff’s Current Law applied to the node connecting R1, R2 and R3.
Call this node “x” with potential Ux:
Ok v Ol v Om y +

U w e + w e wV w e
v v y+
U v > @ V v ?
Solve for Ux:
U e e V e
w w w w y+
U v > U v > @ V v ? V v ?
, , , U V
v v e y w
U v > @ V v ? U v > V v ?
U V
U v > w V v ?
e y z w-"+4;
, , ,
v v
U v > @ V v ?
l h h hR
Now we can calculate the current through R3:
Om e
Om y y z w,/"+
@ @
Where the minus sign means that the current is away from Ux (from the top to
the bottom).
Question 4.2 Equivalent circuit of a power supply

a. A generic Thévenin equivalent circuit is given in Figure 28.
48
Figure 28: Thévenin equivalent circuit
The Thévenin equivalent source can be found by determining the open loop voltage:
the ideal multimeter indicates 10.00V, so UTh = 10.00V.
The Thévenin internal resistor can be found by interpreting the set-up as a voltage
divider with a resistor of 100Ω. We find:
?
adc y ]`
> v ?
,++& ,+ w 4"32
4"32 y ,+ : > y ,++& z ,".-&
> v ,++& 4"32
Join the best at Top master’s programmes

• 3
3rd place Financial Times worldwide ranking: MSc
the Maastricht University International Business
• 1st place: MSc International Business
School of Business and • 1st place: MSc Financial Economics
• 2nd place: MSc Management of Learning
Economics! • 2nd place: MSc Economics

• 2nd place: MSc Econometrics and Operations Research
• 2nd place: MSc Global Supply Chain Management and
Change
Sources: Keuzegids Master ranking 2013; Elsevier ‘Beste Studies’ ranking 2012;
Financial Times Global Masters in Management ranking 2012
Maastricht
University is
the best specialist
university in the
Visit us and find out why we are the best! Netherlands
(Elsevier)
Master’s Open Day: 22 February 2014
www.mastersopenday.nl
49
b. First of all: RTh = RN = 1.32Ω. IN = UTh/RTh | 7.89A

c. P = U·I = U2/R = (9.87V)2/100Ω | 0.974W
Question 4.3 Norton and Thévenin source transformations

a. Drawing: same as Figure 28. The Thévenin equivalent source can be found by
calculating the open loop voltage:
@
Q\ y JK y U y -"0
> v @
and the Thévenin internal resistor is the resistance between node A and B when
the internal source Ua is set to 0V:
> @
Q\ y ? v y -0+&
> v @
Q\
b. M y y ,+
Q\
M y Q\ y -0+&
c. Drawing: same as Figure 28. The Thévenin equivalent source can be found by
calculating the open loop voltage:

and the Thévenin internal resistor is the resistance between node A and B when
the internal source Ua is set to 0V:
0V:
? > v @
Q\ y z -,"4&
? v > v @
d. The circuit is loaded with a 1kΩ resistor:
JK $R`_aUXYX y Q\ y ,",-0
LaUX
JK $LaUXYX y Q\ z ,",+,
Q\ v LaUX
The reduction of the output voltage of 24mV is 2.4%
50
Question 4.4 Equivalent circuit of a battery

a. First of all, the UTh is equal to the open loop voltage 8.5V. Next, RTh can be
found by observing the circuit as a voltage divider as shown in Figure 29.
Figure 29: Lightbulb connected to the Thévenin equivalent of a battery
L][\cVd_V ,++&
3"-0 y Q\ y 3"0
L][\cVd_V v Q\ ,++& v Q\
3"-0
,++& , w 3"0
Q\ y z /,"4&
3"-0
3"0
Question 4.5 Superposition

a. Evaluate two circuits: one with Ua = 0V and one with Ub = 0V.
Figure 30: Partial circuits with Ua = 0V (left-hand) and Ub = 0V (right hand)
51
With a voltage source, “0V” means a short circuit:
> @
> v @
T $Rq I=S y V z ,"/-4
> @
? v v
> @
?@
? v @
T $Rr I=S y U z -".3,
@
> v ?v
? @
T y T $Rq I=S v T $Rr I=S z ."3,+
b.
Ok v Ol v Om y +

U w e V w e + w e
v v y +
> ? @
U V
> v ?
e y z ."3,+
, , ,
v v
> ? @
52

a. The left branch with R1 and R2 is 5V (half of the power supply because R1 =
R2). The right branch is a voltage divider with R3 and RT and is
@
O][\c y U z /"4/.
Q v @
So the difference is
LYZc w O][\c z 0 w /"4/. z 01"3
b. In balance when RT = 1kΩ:
Q
Q y /"4. 5 ,+D 5 H?D"= y ,&
,&
y w-2"+ 5 z -4,"20 z ,3"16
/"4. 5 ,+D
c. Half of the supply voltage: 5V, or (5.000V+4.943V)/2 = 4.9715V exactly

d. By re-using the results of question a) and b):
fR BC"EjiH=i
y y z84%6
'
fQ ?F>">BhH?F>"DBh
Question 4.7 Time dependent circuits

a. The final potential is Ua = 5V
b. W = R·C = 1kΩ·10µF = 10ms
c. Maximum potential difference over R, so I = Ua/R1 = 5mA
d. dU/dt = 0, so I becomes 0mA
e. No change: Ua = UC = 5V, IC = 0mA
Question 5.1 Terminology

a. Sensitivity
b. Saturation
c. Offset or systematic error
d. Drift
e. Resolution
f. Cross-sensitivity
53
g. One-point calibration
h. Differential measurement
i. Common-mode rejection ratio (CMRR)
j. Bandwidth
Question 5.2 The transfer function of a Pt100

a. Draw a line in Figure 11 or use Excel to fit the curve over the range of
0°C-100°C. With ΔR = 230Ω and ΔT = 600°C, we find:
%
y y +".3&#9
%
< y100&
Note that the accuracy based on this graphical method of determining a and b will
be poor. Based on the given raw data, a better fit could be made.
b. The sensitivity is a = 0.38Ω/°C
c. The offset is b = 100Ω for x in °C, which means the sensor is 100Ω for T = 0°C
d.
Figure 31: Fit the line over the range 0-100°C, read the slope out over a larger range
54
Question 5.3 A resistive sensor

a. The sensitivity can be read from the fitted line: dy/dx = -0.0656 MΩ/mm
= -6.56·10-7 Ω/m.
b. Over the range from 0 to 5.5 mm deflection, the non-linearity can be found
from the maximum deviation which is about 0.3 of a vertical unit (MΩ). SO
the non-linearity is 0.3 MΩ/5.5 mm · 100% =
Question 5.4 The candle clock

a.
Figure 32: Burning of the candle with a fitted line
b. The value for t = 0 minutes is equal to b, this is where the line crosses the
vertical axes. So b = 39.65 cm. The slope of the fitted line is parameter a and is
-0.1972 cm/min: almost 2 mm per minute. In SI units this is a = -32.9 µm/sec.
c. The standard deviation is a good measure for the error. In this case it is the
standard deviation of the measured instantaneous candle lengths with respect to
the estimated length at that time using the fitted line. Take the sum of these error
squared, divide by the number of points minus one, and take the square root.
d. Equal to the absolute value of the slope “a”: 32.9 µm/sec.
55
Question 6.1 From sensor to knowledge

a. When signals x and y have the same common error signal, this error will be
removed when the difference between x and y is taken. If x = x’+e and y = y’+e,
then x-y is equal to x’-y’ and will not be sensitive to e. A practical example is
the use of an inaccurate spirit level to measure whether a construction project
is absolutely horizontal. If we use the spirit level twice, where the second
time it is rotated 180 degrees, a systematic error in the spirit level is removed.
Other examples are in sensor systems where we want to cancel out temperature
interference: take a second indifferent sensor that is not measuring the quantity
of purpose, but only the cross interference of temperature.
b. In a stimulus-response measurement we measure how a system responds to a
known actuator input. For example when tapping on a melon to hear whether
there is a high water pressure inside, an acid-base titration, etc.
Question 6.2 Differential measurements

a.
!> y > v > v '

}
!? y ? v ? v '
!? w !> = y ? v ? v ' w > = v > v '
!? w !> = y ? v ? v ' w > = w > w '
!? w !> = y ? w > = v ? w >
Assuming that a1 ≈ a2, this is an output signal where is completely eliminated. The
output y2-y1 is a signal proportional to x-x0.
b.
! y > v ' v >

} >
!? y ? v ' v ?
!? w !> = y ~? v ' v ? w ~> v ' = v >
!? w !> y ? v ' v ? w > = w ' = w >
!? w !> y ? w > = v ? w > v ' w =
( ) df h d ff l l
The last term (x-x0) is not removed from the differential sensor signal.
c. The left and right branch of a Wheatstone bridge are in fact the two inputs
of a differential set-up. As an amplifier to measure a difference, the differential
OpAmp amplifier of Appendix B of the book can be used, or a dedicated
instrumentation amplifier.
56
Question 6.3 A resistive measurement bridge for a train gauge

a. There are several options, but the resistors on one side (“View A” or “View B”)
must be diagonally opposite. For bending downwards, R1 and R3 will become
larger and R2 and R4 smaller. Therefore, in all valid solutions, there is a loop
R1 Æ R2 Æ R3 Æ R4 Æ R1. Only then the differential signal is the result of
bending in the positive or negative x direction.
Figure 33: Correct positioning of the strain-gauge resistors in the Wheatstone

bridge
Need help with your

dissertation?
Get in-depth feedback & advice from experts in your
topic area. Find out what you can do to improve
the quality of your dissertation!
Get Help Now
Go to www.helpmyassignment.co.uk for more info
57
b. The response will be an s-shaped curve. Around x = 0, the second derivative is zero,
meaning this is a deflection point where the curve is linear locally. In practice, for
small changes of the resistances of the gauges, a very high linearity is seen.
Figure 34: An s-curve is linear around x=0
c. With the full bridge of Figure 33, the temperature effect is minimize. If R1, R2,
R3 and R4 are designed with the same nominal resistance (resistance at the same
temperature), the temperature effect is common and equal. The bridge cancels
these effects out.
d. The differential signal is -5mV..+5mV and the common signal is half the power
supply Us/2 = 5V. This means that the required Common-Mode Rejection Ratio
must be better than 10mV/5V = 0.002. The amplifier should amplify 1000x to
map the differential signal onto the AD converter range.
e. Not needed: the strain gauges are very close to each other, and the wires to the
voltmeter are not ideally not carrying a current.

a. The LDR must be 10kΩ to make the left branch equal to the right branch
b. The theoretical linearized sensitivity of a single bridge is
%P S ,+
zw y z +"-0#&
% / = / 5 ,+&
assuming all resistors are 10kΩ. In this case the sensitivity is positive because the output
increases with increasing resistance of the LDR. Since not all resistors are 10kΩ, we
can also calculate the output for 10kΩ and for 10kΩ+1Ω and determine the difference:
58
P $OI>=^g y +
-+& -+&
P $OI>=^gG>g y ,+ w ,+ z +"--#&
,+& v -+& ,+& v ,& v -+&
The differential voltage between the right and left branch is about 0.22mV/Ω,
which is slightly less sensitive than the first estimation.
Question 7.2 An LDR in a voltage divider

a. The transitions are 100 and 1000 lux, that means from 10 to 10000 lux is more
than sufficient to see al events.
b. The range from 1.0 to 10000 lux is equivalent to 100MΩ down to 0.1kΩ, from
10 to 10000 lux is equivalent to 20MΩ down to 0.1kΩ.
c. We want to have hysteresis to avoid a transition state where the screen goes up
and down within seconds. When the two decision levels are switched, the screen
will be stuck to one of the sides.
d. The middle between 100 and 1000 lux is roughly at 2kΩ: that is the correct
resistor.
e. The light level of 100 lux gives about 3kΩ, the level of 1000 lux will be 0.7kΩ.
With the series resistor of 2kΩ this gives transition levels of 3/5 = 0.4 times the
power supply and 0.7/2.7 = 0.3 times the power supply. With a Schmitt trigger
type of comparator, these two levels can be realized.
Question 7.3 Sensor characteristics of an NTC

a. The voltage divider is probably more non-linear than the sensor over
that small range.
b. To sense Utemp without loading the voltage divider resistively.
c. Two degrees is 0.03·0.03 = 0.0009, so the resistance becomes 970Ω·(1-0.0009)
= 969.1Ω. It gets lower with increasing T because it has a negative temperature
coefficient. Utemp (and so Uout) will go up from 5V·1000Ω /1970Ω ≈ 2.5381V to
5V·1000Ω /1969.1Ω ≈ 2.5392V. This is a difference of only 1.13mV on top of
an offset of about 2.5V.
d. Make a Wheatstone bridge with more resistors of 1kΩ to reduce the
common signal so the differential voltage of 1.13mV becomes relatively larger
and can be amplified.
59
e. As long as the temperature dependency of R2 is not equal to that of the

sensor R1, the voltage divider ill still give a voltage change with changing
temperature, only smaller. The lowered change of voltage can be registered
with calibration, and if desired removed by increasing the amplification factor
of a differential amplifier.
Question 7.4 Schmitt trigger

a. The OpAmp is a comparator where the decision level Uc is dependent on the
history. For a signal going from low to high, Uc is higher than a signal going
from high to low.
b. If Uout is initially low (0V), the circuit is effectively as in Figure 35a, when Uout
is high (5V), the circuit is effectively as in Figure 35b.
Brain power By 2020, wind could provide one-tenth of our planet’s

electricity needs. Already today, SKF’s innovative know-
how is crucial to running a large proportion of the
world’s wind turbines.
Up to 25 % of the generating costs relate to mainte-
nance. These can be reduced dramatically thanks to our
systems for on-line condition monitoring and automatic
lubrication. We help make it more economical to create
cleaner, cheaper energy out of thin air.
By sharing our experience, expertise, and creativity,
industries can boost performance beyond expectations.
Therefore we need the best employees who can
meet this challenge!
The Power of Knowledge Engineering
Plug into The Power of Knowledge Engineering.

Visit us at www.skf.com/knowledge
60
Figure 35: In a Schmitt trigger, one of the resistors determining the decision level of the comparator is
effectively moved in place
For Uout = 0V:
?@
? v @
W $RsutI=S y WW y +"4;
?@
> v v
? @
For Uout = 5V:
?
W $Rsut IBS y WW y ,"2
> @
> v @ v ?
h h k d
These are two equations with three unknown resistors R1, R2 and R3. Take for example R1
= 10kΩ, and we find:
?@
? v @
0 y +"4;
@
,+& v ?v
? @
?
0 y ,"2
,+& @
v ?
,+& v @
Which can be solved as R2 = 2727.27Ω and R3 = 11250.0Ω.
61
Question 8.1 AD Conversion

a. The resolution is defined as:

y z
-M w , -M
So:

>=
| ? y y 4"42
>= -
Which means that N = 10 to make the outcome larger than 9.97. And indeed,
with 10 bits over a range of 5V, we find a resolution of 4.89mV which is better
than 5mV.
b. The Nyquist theorem says a sampling rate is needed which is at least higher than
two times the highest frequency in the signal. So, this is 2kHz. Only then, the
original analog signal can be resolved from the digital samples.
Question 8.2 Measurement of light intensity

a. 5 lux Æ 74kΩ, 50 lux Æ 15kΩ,
b. No, when the transfer curve looks linear on a log-log scale, it must be strongly
exponential on a linear scale. There is no single slope for the given range.
c. A voltage divider, for example as given in Figure 36.
Figure 36: A voltage divider can be used to convert the value of the resistive
temperature sensor into a voltage
62
d. The value for R1 can be chosen in the middle of the operational range of the
sensor: 47kΩ for example. With Ua = 5V, the output voltage will be between
1.92V and 3.75V for the range from 5 to 50 lux.
e. An AD converter with a range of 5V and 12 bits has a resolution of 5V/(212-1) 
| 1.22mV, so this is not enough for detecting 1mV variations.
f. The day-night variation of the light intensity is probably not a sine wave. We
are interested in the exact transition from dark to light or the other way around.
That signal if faster than a period of 24 hours. Normally, there is no big issue
with sampling every minute.
Question 8.3 A temperature measurement

a. For measurement around room temperature the NTC has the largest variation.
In numbers: from 10 to 25°C the Pt100 changes from 107Ω to 119Ω (11%)
while the NTC changes from 590Ω to 350Ω (28%). This is much easier to
work with in a voltage divider. However, when a high linearity is needed, the
Pt100 is better.
63
b. See Figure 36 for the circuit. When the NTC is used, R1 becomes 400 Ω
because that is the sensor value at the room temperature of 20°C.
c. If the non-linearity over the needed range is more than accepted, a
microcontroller can be used to calculate the temperature from the instantaneous
sensor signal using a calibration curve.
Question 9.1 Bus interfaces

a. (1) Digital signals are less susceptible to noise, (2) Standardization, (3) all sensors
can be placed on the same bus, (4) meta data for identification and calibration
can be sent over the same bus, (5) reduction of the number of wires.
b. Standardization, low noise, simplified maintenance, error checking, internet over
the same bus.
Assignment 1: The position sensor
Question 10.1 First measurement – Does it work?

a. The experience is that with the given materials and sizes, the resistances are in
the tens of MΩ’s.
Question 10.2 Calibration curve

a. An example is given in Table 3.
64
d [cm] R [MΩ]
10 12,07
9 11,53
8 11,15
7 10,52
6 8,92
5 7,24
4 6,02
3 0,524
2 0,274
1 0,277
Table 3: Resistance measurements using a Voltcraft VC170 multimeter in auto
b. The values of Table 3 are plotted in Figure 37. The linear fit and the error bars
will be explained later.
Figure 37: Measured resistances of the position sensor
65
Question 10.3 Analyzing

a. The dependency is clearly linear from d = 4cm.
b. Yes, we are measuring a homogeneous strip ( U is constant) with a fixed with and
a changing length L.
c. The resistance is somewhat high. We can make it lower by adding more carbon
or by using a wider strip.
d. The values below d = 4cm cannot be trusted: we expect a linear behavior and
these are obviously not on the curve.
e. The sensitivity is determined in Figure 37, but only starting from the fourth
point. The first three points, for distances smaller than 4cm, are seen as outliers.
The fitted line gives a slope of 1.0343MΩ/cm. So the sensitivity is 1.0MΩ/cm,
and there is an offset of 2.4MΩ.
f. The slope “a” is the linearized sensitivity of the sensor. It is the first
derivative of the resistance R on the distance d (or length L of the resistor).
The resistor equation is

y*

so
7 *
y y
7
g. The offset is the resistance for L = 0cm. This can be the resistance of the cables
plus the contact resistance of the paperclip to the paper.
Question 10.4 Error estimation

a. To obtain better insights, the experiment is repeated many more times. This is
summarized in Table 4.
b. Also in Table 4.
66
Second measurement: error analysis
d = 2cm d = 6cm d = 10cm d = 2cm d = 6cm d = 10cm
R [MΩ] R [MΩ] R [MΩ] R [%] R [%] R [%]
0,274 8,92 12,07
0,159 18,91 37,82 100,0% 100,0% 100,0%
0,188 17,99 37,91 118,2% 95,1% 100,2%
0,235 18,75 38,91 147,8% 99,2% 102,9%
0,204 24,45 40,23 128,3% 129,3% 106,4%
0,287 24,24 40,01 180,5% 128,2% 105,8%
0,410 19,38 34,47
Mean [MΩ] Mean [MΩ] Mean [MΩ]
0,215 20,868 38,976
St.Dev [MΩ] St.Dev [MΩ] St.Dev [MΩ]
0,049 3,194 1,131
45,6% 30,6% 5,8%
Table 4: The position experiment repeated
c. The data of Table 4 is plotted in Figure 37. With this data we can determine
standard deviations in order to plot error bars in Figure 37. Now we are allowed
to say that the fitted linear approximation fits definitely within the error range of
the measurements, again, only for distances above 4cm.
The repeated experiments are also plotted in Figure 38. We can see that over time,
the sensor value is drifting.
67
Figure 38: The sensor value is drifting when the experiments are repeated
Question 10.5 Repeat calibration experiment

a. The characterization (calibration) of the sensor as was done in Figure 37, is
repeated after all experiments of Table 4 are done. The result is the calibration
curve Figure 39.
b. See Figure 39.
Figure 39: Sensor characterization when the experiment is repeated
68
Question 10.6 Analysis

a. There are obviously two types of drift. First of all, the sensitivity (slope “a”) was
1 MΩ/cm, and after some use it became 3.7 MΩ/cm. An explanation can be
that the carbon becomes crunched and mixed. If that is the case, the drift will
stop because this process will stabilize. Besides drift in the sensitivity, there is
also drift in the offset: this was 0.7MΩ, and became 3.7MΩ. An explanation of
a change in the offset can be a change in the contact resistance.
b. Answered under a): it is probably the grain size of the carbon that is changing.
c. In fact, the measurement error of this experiment is much smaller than the drift.
The drift is that large that we must conclude the sensor cannot be calibrated in
a useful number of x MΩ/cm.
69
Assignment: Strain gauge transducer
Question 11.1 First measurement – Does it work?

a. Figure 40 shows the set-up where a plastic plate is coated on one side with a
carbon pencil. The resistance is continuously measured with a multimeter. The
bending can be measured with a ruler drawn on a piece of paper.
Figure 40: The set-up for measuring a bending plastic plate
b. The measurements are in Table 5, and are plotted in a graph in Figure 41 for
positive numbers. Positive numbers in the deflection are defined here as bending
the plastic strip downwards. Using Excel’s fit function, a line is fit in the plot.
The slope is -0.0626 MΩ/mm.
d [mm] R [MΩ]
0 2,17
1 2,07
2,5 1,98
3 1,95
5 1,85
-1 2,16
-2,5 2,15
-5 2,14
-7,5 2,21
-10 2,3
Table 5: Measured resistances at several bending distances
70
Figure 41: Calibration curve of the strain gauge for a positive (downwards) deflection
c. It is linear. One quantitative evidence for that is that the value of R2 is very
close to one: 0.98.
d. When bending upwards, we can see a strange phenomenon. While Figure 41
showed only the positive values of Table 5, the negative numbers (bending
upwards) are included in Figure 42 as well.
Figure 42: Bending the strain gauge in two directions being upwards and downwards
71
It looks like bending upwards gives a much lower sensitivity: there is less deformation of
the resistive layer than when bending downwards. The drawing in Figure 43 explains what
may be the cause in the used set-up: the radius of curvature differs depending on the
bending direction.
Figure 43: Bending upwards and downwards give two different effective lengths of the strain gauge
Question 11.12 Read-out electronics

a. A Wheatstone bridge eliminates large sensor offsets. Take the resistor values of
the bridge equal to the nominal value of the sensor. This is about 2.2MΩ.
Figure 44: A Wheatstone bridge to turn the resistive strain gauge into a
voltage with low offset
Question 11.3 The strain gauge

a. Should not deviate more than a factor of 10 from the given value of 3.5MΩ
b. This is 5cm. With a width of 2.5cm and a length of 5cm, a straight non-
meandering configuration makes the most sense.
72
c. Temperature coefficients are defined as
, %
y y w+"+++09H>
%
We allow 5% over a full change of resistance of 1%, this is 0.01∙0.05=0.0005. The

given temperature coefficient of -0.0005 per °C makes that a variation of exactly
1°C results in the maximum allowed error in the resistance.
Question 11.4 Bridge set-up

a. The sensitivity of a single sensor bridge (a quarter bridge) is
%
NdUbcYbKb]X[Y z w
/ = S
and take UV = 5V and R0 = 2.2MΩ. We will find -0.57μ V/Ω.
b. For a 1Ω change in the sensor, the differential signal will be 0.57μV while the
common signal is 2.5 V. This is a ratio of 4.4∙106.
c. The sensitivity will stay approximately -0.57μV/Ω but the rest signal will get an
offset. Part of the advantage of the bridge is removed: it no longer removes the
complete offset from the sensor signal.
Question 11.5 Half bridge set-up

a. The sensitivity is doubled
b. In the ideal case, no difference
73

Electronic Measurements Exercises and Assignments

Uploaded by

Copyright:

Available Formats

Electronic Measurements Exercises and Assignments

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Electronic Measurements Exercises and Assignments

Uploaded by

Copyright:

Available Formats

﻿

3 Electric Currents and Potentials 9

5 Basic Sensor Theory 16

7 Signal Conditioning and Sensor Read-out 25

8 ADC and DAC 31

10 Assignment 1: The position sensor 35

11 Assignment: Strain gauge transducer 39

Question 1.1 SI Units and conversions

j. Express the Ohm in SI units

Question 1.2 The measurement chain

Figure 1: A fever thermometer

e. What type of measurement scale is the temperature indicator?

Question 2.1 T﻿he USB adapter

4.821 V 4.902 V 4.919 V 4.782 V 4.770 V

a. Calculate the mean

Question 2.2 Error propagation and representation

Both have a 5% tolerance

Question 3.1 Electric Currents and Potentials

a. What is the resistance of the described cable per km?

Question 3.2 AC Currents

Question 3.3 Resistors in series and in parallel

Figure 2: A network of resistors

Figure 3: A network of resistors

Figure 4: A network of resistors

Question 3.4 The voltage divider

Question 4.1 Kirchhoff’s laws

Figure 5: An electronic network of voltage sources and resistors

Question 4.2 Equivalent circuit of a power supply

Question 4.3 Norton and Thévenin source transformations

Figure 7: A second network

Question 4.4 Equivalent circuit of a battery

Question 4.5 Superposition

Figure 8: A network with two voltage sources

a. Calculate UX of the network in Figure 8 using superposition.

Question 4.6 The Wheatstone bridge

© Deloitte & Touche LLP and affiliated entities.

Deloitte & Touche LLP and affiliated entities.

Discover the truth at www.deloitte.ca/careers

Take Ua = 10V, R1 = R2 = R3 = 1kΩ.

a. At room temperature (18°C) the value of RT is equal to 1023Ω. What indicates

Question 4.7 Time dependent circuits

Figure 10: An RC circuit

5 BASIC SENSOR THEORY

Question 5.1 Terminology

Question 5.2 The transfer function of a Pt100

Figure 11: Calibration curve of a Pt100 temperature sensor

Some raw data is available as well in Table 1.

We will turn your CV into

Table 1: Measurement points form a Pt100 sensor

a. Fit a linear model (a·x+b) for the range 0°C to 100°C

Question 5.3 A resistive sensor

Figure 12: Measured calibration curve of a strain gauge

Question 5.4 T﻿he candle-clock

Figure 13: A measurement setup for a candle-clock

Figure 14: Measured calibration curve of the candle clock

Question 6.1 From sensor to knowledge

Question 6.2 Dif﻿ferential measurements

Question 6.3 A resistive measurement bridge for a strain gauge

Figure 15: Configuration with four strain gauges on a lever

Question 2.1 The USB adapter

Question 5.4 The candle-clock

Question 6.2 Differential measurements

Question 11.3 The strain gauge

/"3-, w /"3./? v /"4+- w /"3./? v /"4,4 w /"3./? " "